At 18, I shipped a production SaaS billing platform as the sole developer at a US-based AI startup —
95+ API endpoints, end-to-end Stripe integration, a 15+ page Next.js admin portal, and a fully normalized
Azure MSSQL database. Backend, frontend, database design — all of it, independently.

• 💼 Full-Stack Developer at TruBotAI USA LLC — building subscription management systems across Node.js, Next.js, TypeScript, and MSSQL
• 🐍 5 years of Python — from terminal automation and CLI tools, through KivyMD GUI apps, to production Django and Flask backends
• 🔁 Comfortable switching stacks — Python is where I think, Node.js is where I currently ship
• 📍 Based in India · Open to relocation anywhere

I don't just write code — I think through systems before touching a keyboard.

When I joined TruBotAI with no prior professional experience, I was handed a SaaS product to build from scratch. No existing backend. No database. No frontend. The requirements were complex — multi-product subscriptions, Stripe billing, IP-based pricing, multi-currency support, and a full admin portal. Here's how I approached it:

• Database first — I designed 16 relational tables before writing a single API, mapping out foreign keys, normalization, and indexing strategy upfront to avoid painful migrations later
• Modular architecture — split the backend into independent modules (auth, billing, subscriptions, admin, webhooks) so each could be built, tested, and debugged in isolation
• Stripe webhooks — instead of polling for payment status, I architected a full event-driven webhook system that keeps billing state in sync in real time with zero manual intervention
• Background jobs — identified that renewal processing and plan expiry were time-sensitive operations that shouldn't block the request cycle, so I implemented node-cron jobs to handle them asynchronously
• Query optimization — when subscription and invoice queries slowed down under realistic data volumes, I profiled the queries, identified missing indexes, and applied targeted fixes rather than rewriting logic

This is the same thinking I bring to every project — understand the problem deeply, design before building, optimize where it actually matters.

Django REST Framework React PostgreSQL Django Channels Redis WebSockets Docker

This project started with one question: how does real-time actually work? I didn't want a tutorial answer — I wanted to build it from scratch and understand every layer.

The problem-solving behind it:

• Hybrid architecture — used HTTP for standard REST operations and WebSockets only where real-time was actually needed, keeping the system clean instead of over-engineering everything as WebSocket
• Redis channel layers — when I realized Django Channels alone wouldn't scale beyond a single process, I introduced Redis as the message broker so multiple workers could share state
• Auth complexity — implemented three layers: JWT for API auth, OAuth for social login, and TOTP-based 2FA via authenticator app — each solving a different threat model
• Production deployment — moved beyond runserver and learned Daphne (ASGI server), Nginx as reverse proxy, Docker Compose with named volumes for persistent storage, and how to wire them together correctly

Result: A fully containerized, production-grade real-time chat platform with a complete auth stack — built purely to understand the technology deeply.

Flask PostgreSQL HTMX SQLAlchemy Docker Nginx Gunicorn GitHub Actions

Built as a Medium-inspired blogging platform — but the interesting part wasn't the CRUD, it was the decisions made around content management, performance, and deployment.

The problem-solving behind it:

• Custom Markdown editor — instead of dropping in a third-party library, I built an editor toolbox that auto-inserts Markdown syntax (headings, code blocks, inline code, images) because I wanted full control over the writing experience
• Image asset management — designed a system that accepts uploads, stores them, and returns optimized Markdown-ready public URLs — so writers never have to think about file paths
• HTMX for interactivity — chose HTMX over a full React frontend because the interactivity requirements (nested comments, partial updates) didn't justify the complexity of a SPA. This kept the stack simple without sacrificing UX
• SQLAlchemy lazy loading — identified N+1 query problems early using query profiling and resolved them with lazy loading, pagination, and indexed queries before they became production issues
• Full CI/CD pipeline — set up GitHub Actions to automate testing, linting, and deployment so the feedback loop stayed tight

Result: A production-ready platform containerized with Docker Compose, served behind Nginx + Gunicorn, with automated deployments on every push.

Languages     Python · TypeScript · JavaScript · SQL · C++ · Java
Backend       Django · DRF · Flask · FastAPI · Node.js · Express.js
Frontend      React · Next.js · Tailwind CSS · HTMX · Bootstrap · SCSS
Database      PostgreSQL · MSSQL · MongoDB · Redis · SQLAlchemy
DevOps        Docker · Nginx · Gunicorn · Daphne · GitHub Actions · Azure · Linux
Auth          JWT · Session Auth · OAuth · 2FA (TOTP)
Concepts      REST API Design · System Design · WebSockets · Query Optimization
              Caching (Redis) · Indexing Strategies · Modular Architecture

My Python journey didn't start with web development. It started with.

2020  →  Terminal scripts, automation, CLI tools
2021  →  Python libraries — pytube, spotdl, ffmpeg, mutagen, Pillow
2022  →  Kivy & KivyMD — cross-platform GUI desktop + mobile apps
2022  →  First Django app — role-based e-commerce site
2023  →  Flask, PostgreSQL, SQLAlchemy — proper backend architecture
2023  →  DSA — Java
2024  →  Memory management and pointers — C++
2024  →  Django REST Framework, React, real-time with Django Channels
2024  →  Docker, Nginx, Gunicorn — production deployments
2025  →  FastAPI, Redis, 2FA, OAuth, CI/CD with GitHub Actions
2025  →  Node.js, TypeScript, Next.js — production at TruBotAI

Every step was driven by a problem I wanted to solve, not a curriculum I was following.

• 🥇 First Place — Sub-district Level Web Design Competition (2024)
• 🥇 First Place — Sub-district Scratch Programming (2023)
• 📊 99% in SSLC (2022–2023) · 92% in Higher Secondary Computer Science (2023–2025)
• 🏭 Shipped to production at 18 as sole developer at a US-based AI SaaS startup
• 🎓 Self-taught — no degree, all built through curiosity and real projects

I believe the best way to understand something is to build it, break it, and rebuild it better.
Every project in this profile exists because I had a question I needed to answer through code.